--- dataset_info: features: - name: image dtype: image - name: ground_truth dtype: string splits: - name: train num_bytes: 450788921.0 num_examples: 695 download_size: 313374577 dataset_size: 450788921.0 configs: - config_name: default data_files: - split: train path: data/train-* ---